Aging of the skeleton is associated with increased prevalence of osteocyte apoptosis in mice and humans. Work leading to this application shows that the skeletal syndrome that ensues with aging is mimicked by osteocyte specific deletion of the gap junction protein connexin (Cx) 43. Mice lacking Cx43 in osteocytes exhibit increased osteocyte apoptosis and accumulation of empty lacunae, and reduced OPG levels, and Cx43- deficient osteocytes express lower levels of the survival-associated microRNA (miR) miR21. In addition, bones from old mice exhibit reduced Cx43, OPG, and miR21 expression. We showed in vitro that apoptotic osteocytes release high mobility group box protein1 (HMGB1), which reduces OPG expression and increases osteoclast recruitment and differentiation through activation of receptors for advanced glycation end products (RAGE). HMGB1 also activated toll-like receptor 4 (TLR4). However, whether activation of HMGB1 receptors RAGE or TLR4 has a role in the skeletal phenotype of Cx43-deficient or old mice is not known. The mechanism by which HMGB1 is released by apoptotic cells is not known. A potential mechanism could involve opening of Pannexin1 (Pax1) channels in cells undergoing apoptosis. Whether this mechanism operates in osteocytes undergoing apoptosis is not known. The long-term goal of our studies is to improve the management of the adverse skeletal effects of aging by targeting osteocyte apoptosis and resorption. The specific objective of this application is to define the signaling pathways involved in the elevated osteocyte apoptosis and bone resorption in aging. We propose that the bone fragility syndrome that ensues with aging is due to increased osteocyte apoptosis as a consequence of reduced expression of osteocytic Cx43, OPG, and miR21. We hypothesize that low OPG and the consequent increase in the RANKL/OPG ratio acts as a permissive event for osteoclast development; whereas low miR21 reduces anti-apoptotic kinase signaling and activates caspases, with the consequent Panx1 channel opening and HMGB1 release, directing osteoclasts to areas with accumulated apoptotic osteocytes. To test this hypothesis, we will pursue a combination of in vivo and in vitro studies that include the use of novel genetically modified mice, and established and primary osteocytic cells. Aim 1 will examine the role of Cx43/miR21 on osteocyte apoptosis with aging. Aim 2 will investigate the role of HMGB1 in osteoclast recruitment and differentiation in Cx43 deficiency and aging. Aim 3 will explore the consequence of osteocytic Panx1 removal on HMGB1 release and osteoclast recruitment. Successful completion of these studies will provide the molecular basis for the in- creased osteocyte apoptosis in the absence of Cx43 and in old animals. In addition, these studies will establish the molecular link between dying osteocytes and targeted bone resorption. Moreover, they will offer the basis for treatments in which the effect of increased osteocyte apoptosis on osteoclast recruitment are counteracted by targeting the HMGB1-RAGE/TLR4 system.